Netflix-RSS Example with Kieker

As the netflix-RSS example consists of currently three components (netflix-RSS-eureka, netflix-RSS-middletier, netflix-RSS-edge), we need some way to interconnect them in the docker context. To achieve this, we use Docker Compose (https://www.docker.com/docker-compose).

Docker Compose makes it easy to pull, start and interconnect several Docker containers by just providing one "recipe". In this repository we provide a recipe to set up the netflix-RSS example and this short introduction on how to use it.

We used the inspectIT Docker setup as a reference to set up an example that is similar to theirs but with Kieker instrumentation instead of inspectIT.

How to set it up

First of all we need Docker and Docker Compose for this to work:

• Docker: https://www.docker.com
• Docker Compose: https://docs.docker.com/compose/install/

Now you can use the docker-compose.yml file in this repository to create the netflix-RSS example as follows. Note: To have this example working, the ports 8080and 9090 should be free. Otherwise you have to change the port mapping in the docker-compose.yml.

• Open a terminal and change to the directory where the docker-compose.yml file is.
• Execute the command: sudo docker-compose up (if you are root, you can leave out the sudo)

Now Docker should pull the containers if they are not present and start the netflix-RSS example.

To test the example application on your machine, you can open the URLs:

1. http://localhost:8080/eureka
2. http://localhost:9090/jsp/rss.jsp

When the system has started properly (usually after 2-3 min) you can add RSS links to the input field on the second site (http://localhost:9090/jsp/rss.jsp) and see them in the listing below.

How to use this example for analysis

As the containers are instrumented with Kieker, the monitoring logs can be found in /tmp/netflix-rss/{edge|eureka|middletier}/logs. This data can be used for a trace analysis for example.

Trace analysis with Kieker

To be able to do a trace analysis you have to download and extract the Kieker release (http://kieker-monitoring.net).

This is an example of what you could do with the data. For further details on what else you can do with the monitored data, please refer to the Kieker documentation (http://kieker-monitoring.net/documentation/).

Steps:

1. Change into the bin folder in the extracted Kieker release folder.
2. Create a folder where the output file should be stored (in this case we assume this to be /tmp/netflix-rss/trace-analysis)
3. Execute ./trace-analysis.sh -i $(find "/tmp/netflix-rss" -type d -name kieker-*) -o "/tmp/netflix-rss/trace-analysis" --plot-Deployment-Operation-Dependency-Graph --ignore-invalid-traces to do the trace analysis.
4. To convert the analysis data to PDF format, issue the command: ./dotPic-fileConverter.sh /tmp/netflix-rss/trace-analysis pdf
5. Now there should be a PDF file representing the plotted monitored traces in the /tmp/netflix-rss/trace-analysis pdf folder.

Customizing the monitoring

If you want to change the configuration for the monitoring right from the beginning you can provide your own configuration files (kieker.monitoring.properties, aop.xml). To do so, place them in the following directories:

• /tmp/netflix-rss/{edge|eureka|middletier}/config/kieker.monitoring.properties
• /tmp/netflix-rss/{edge|eureka|middletier}/config/META-INF/aop.xml

These files will replace the default configuration.